The goals of this project are a) to develop and refine parent report and observational assessment techniques for measuring infant temperament, and b) to utilize these instruments in a short-term longitudinal study of infant temperament and caretaker behavior. We define temperament as relatively enduring individual differences in activity level, emotionality, and deployment of attention. Data collection will be carried out for 75 infants, to be seen longitudinally at 3, 6, 9, and 12 months of age. Data will include: home and laboratory structured assessments of temperament-related behavior, parent-reported infant temperament as measured by the Infant Behavior Questionnaire, and measures of caretaker-infant interaction. We will carry out multi-trait, multi-method analyses of our scores, testing for sex and birth order similarities and differences and stability of scores over time, predicting temperament scores from earlier scores on other scales, assessing intercorrelations among scales, and testing for the relation between caretaker behavior and infant temperament.